Standardization of Core Laboratories
Received: March 31, 2004
Accepted: August 16, 2004
Published online: November 18, 2004
Heart Drug 2004;4:201–217
DOI: 10.1159/000082191
Methods for the Evaluation of Vascular
Reconstruction
Einar Stranden
Department of Vascular Diagnosis and Research, Aker University Hospital, Oslo, Norway
Key Words
Arterial reconstruction W Doppler flowmetry W
Electromagnetic flowmetry W Flowmetry W Intravascular
ultrasound W Outflow resistance W Pressure recording W
Transit time flowmetry W Ultrasound
Abstract
Despite carefully performed surgery, vascular defects
can remain undetected within an arterial reconstruction
after blood flow is restored. Failure to recognize technical errors or abnormalities can lead to vessel thrombosis
and embolization. For cardiac, carotid or visceral reconstructions these events can be catastrophic. Furthermore, corrections of lesions before graft thrombosis can
have a significant impact on long-term patency, being
particularly important for vein grafts, because most will
not maintain patency with thrombectomy alone. This
review describes methods for peroperative evaluation of
vascular reconstructions. Since careful selection of patients for therapy as well as proper postoperative surveillance contribute to the success of vascular surgery, some
methods for pre- and postoperative evaluation are also
included.
Copyright © 2004 S. Karger AG, Basel
ABC
© 2004 S. Karger AG, Basel
1422–9528/04/0044–0201$21.00/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Accessible online at:
www.karger.com/hed
Introduction
The need to measure blood flow intraoperatively became apparent during the development of reconstructive
arterial surgery. However, many surgeons unfortunately
still rely on pulse palpation as an index of flow, not realizing that a vessel can pulsate even when there is no blood
flowing through it. In fact, if the vessel occludes distal to
the palpation site, the pulse may even increase, though
admittedly not for long (fig. 1).
The primary aim in flowmetry is to obtain information
on the immediate results of the reconstruction, where a
technical failure may jeopardize an otherwise successful
operation. There is a clear correlation between the blood
flow values obtained following injection of papaverine
and the prognosis of the arterial reconstruction. Dedichen
[1] found in patients with lower limb arterial reconstruction that the risk of early postoperative occlusion was significantly increased if the basal blood flow after reconstruction was less than 100 ml/min or the papaverineinduced flow was less than 200 ml/min. However, it can
be questioned whether the flow measurements are sensitive enough to detect minor technical failures since normal blood flow values have been measured despite extensive defects at the anastomosis [2]. Furthermore, the levels of blood flow recorded during an operation may have
little relation to normal activity. Under some type of anesthesia and conditions due to acute blood loss, blood flow
Prof. Einar Stranden, PhD
Department of Vascular Diagnosis and Research
Surgical Clinic, Aker University Hospital
NO–0514 Oslo (Norway)
Tel. +47 2289 4000, Fax +47 2222 9381, E-Mail einar.stranden@medisin.uio.no
Fig. 1. Pulse palpation is a poor indicator of blood flow. If the artery
is occluded distal to the palpation, the pulse is increased compared to
an open artery. Because the artery is compliant, there is some blood
flow back and forth into the occluded segment. This movement
might produce a Doppler sound interpreted as normal, whereas the
spectrum more correctly shows a reverberant flow with zero mean
blood flow. BP = Blood pressure.
is likely to be markedly depressed, whereas hyperemia is
observed in another type (epidural anesthesia).
The basal blood flow after bypass grafting is of limited
help to precisely evaluate the arterial reconstruction, since
it depends on various factors in addition to the graft itself,
e.g. arteriolar tone, blood volume, body temperature and
the type of anesthesia. The blood flow increase observed
following administration of a vasodilating agent, e.g. papaverine or iloprost [3–5], is a significantly better indicator of the long-term prognosis, and is better suited for the
detection of technical failures.
Five-year patency of femorodistal bypasses varies between 40 and 70% in many series. Some of the reocclusions may be due to intimal hyperplasia, whereas others
are caused by technical failures leading to secondary
thrombosis formation and graft occlusion. To eliminate
technical failures, intraoperative control of the reconstruction is important. Furthermore, such investigations
can also be helpful in planning the procedure, and in giving an indication of the long-term prognosis of the operation.
In cardiac surgery, flowmetry was introduced at an early stage to indicate the severity of disease, and to assess
patency of coronary bypass grafting and the prognosis of
therapy [6–12]. In coronary surgery, flowmetry is still
regarded as an essential tool for evaluating the anastomotic sites [13, 14], and guidelines for performing and interpreting flowmetry procedures have been developed [15].
202
Heart Drug 2004;4:201–217
Fig. 2. A The principle of an electromagnetic flowmeter. e.m.f. =
Electromotive force. B The electromagnetic flow probe for 8-mm
vessels (Nycotron electromagnetic flowmeter). The arrow points at
one of the two surface electrodes.
The purpose of this presentation is to review methods
for evaluation of arterial reconstructions, and to discuss
which methods are going to be important in the future.
Functional Tests
Electromagnetic Flowmeter
Electromagnetic flowmetry quantifies blood flow in a
single vessel. The technique is based on the principles of
electromagnetic induction, described by William Faraday
170 years ago.
If an electrolyte (such as blood) flows at right angles to
a magnetic field, then an electromotive force is induced in
a plane, which is mutually perpendicular to the magnetic
field, and to the direction of fluid flow (fig. 2). Two electrodes situated in the appropriate plane and connected to
Stranden
a suitable detector circuit can measure the induced voltage, which is proportional to the strength of the magnetic
field and to the velocity of blood flow within the blood
vessel. For most clinical and experimental applications,
the probe is C-shaped so that it can be slipped around the
vessel. The probe must be carefully chosen so that it fits
tightly round the vessel to ensure a good contact between
the electrodes and the vessel wall, but does not compress
it. The probe is designed to produce an even magnetic
field, ensuring accurate measurements. Incorporated in
most electromagnetic flowmeter transducers is the electromagnet in addition to the two recording electrodes
(fig. 2B). As the probes are manufactured for specific vessel diameters, a set of differently sized probes is needed.
These are usually calibrated to give blood flow (ml/min).
Most flowmeters utilize an alternating magnetic field to
avoid polarization of the detector electrodes [16].
One important factor is zero stability [17]. When
recording from peripheral vessels occluding the vessel
with a clamp can check the zero reading. However, this is
often impossible when flowmeters are used on the aorta or
pulmonary artery.
Another source of error in making electromagnetic
flow measurements is the variation in sensitivity that is
observed with changing hematocrit [18, 19], influencing
several commercial flowmeters. Consequently, whenever
possible the instrument should be calibrated for the particular patient. The sensitivity to hematocrit is largely due
to the changing electrical impedance of blood, which also
changes when blood is moved from a state of rest, as the
erythrocytes tend to move to the center of the vessel (axial
accumulation). This supposedly leaves a plasma layer at
the vessel wall.
Recording of blood flow in expanded polytetrafluoroethylene (PTFE) grafts with electromagnetic flowmetry is
limited by the electrical isolation of the graft material
[20]. In patients with such graft material, measurements
are usually performed in the native vessel proximal to the
proximal anastomosis or distal to the distal anastomosis.
Ultrasound Techniques
Sound waves are pulsating pressure waves, and when
above audible, e.g. above 20,000 Hz, they are termed ‘ultrasound’.
The sound is characterized by:
E Intensity, i.e. sound pressure or pressure amplitude.
E Frequency, i.e. number of oscillations per second (Hz).
Audible sound is between 20 and 20,000 Hz. Medical
ultrasound is normally ranging between 2 and 15 MHz
(= 2–15 million Hz), although intravascular ultrasound
Methods to Evaluate Vascular
Reconstruction
Fig. 3. Spatial resolution in ultrasound consists of axial and lateral
resolutions. The imaging of point targets results in a smeared-out
image. Note that the lateral resolution is poorer than the axial resolution, and that it is dependent on depth, unlike axial resolution (redrawn from Angelsen [21]).
(IVUS) equipment may use ultrasound frequencies of
30 MHz.
E Wavelength (Ï), i.e. distance between two adjacent
maximal or minimal sound pressure levels.
An important property of the medium where the sound
propagates is the specific ultrasound velocity (c). In air
this is approximately 340 m/s, in water approximately
1,500 m/s. Human tissue is similar to water and the sound
velocity is approximately 1,540 m/s. The wavelength is
the ratio between velocity and frequency (f): Ï = c/f. For
medical ultrasound in human tissue Ï is 0.15 – 0.75 mm
(for audible sound Ï is between 2 cm and 20 m), dependent on the frequency used.
Ultrasound energy is generated by the use of piezoelectric ceramics with electric wires on each side (transducer).
The ceramic is compressed and expanded in parallel with
an applied oscillating voltage, acting like a loudspeaker.
On the other hand, if the ceramic is compressed and
expanded by a sound wave, a variation in voltage is generated, and the ceramic acts as a microphone.
Resolution
Resolution is the ability to detect details. In an ultrasound system there is a clear distinction between axial
(along the sound wave) and lateral resolution (90° to the
sound wave). Axial resolution is usually higher than lateral
resolution. It is dependent on the frequency of the ultrasound emitted (the higher the frequency, the more detailed
is the image obtained), but independent of the depth of the
Heart Drug 2004;4:201–217
203
tissue is the very basis for generating images with anatomical information in B-mode scanners.
Scattering. When a reflecting object is small compared
to the ultrasound wavelength, a portion of the wave energy is scattered in all directions. As scattering is proportional to the 4th power of the frequency, the attenuation
increases steeply with increasing frequency.
Absorption. As the sound wave travels through tissue,
some of the energy is converted to heat by absorption. Absorption is proportional to the 1st power of the frequency.
The selection of a specific frequency for various diagnostic procedures always involves a compromise between
the optimal ultrasound penetration depth (lower frequency – higher penetration depth) and the ability to identify
relevant tissue changes (higher frequency – better sharpness or resolution). For cardiology and abdominal examinations, transducers with frequencies between 2.5 and
3.5 MHz are chosen, for obstetrics 3.5–5.0 MHz, and for
peripheral vessel examinations 5.0- to 15.0-MHz probes
are normally preferred.
Fig. 4. Schematic drawing of a Doppler velocity detector. An ultrasound beam is transmitted by the transducer. Echoes from particles
that have a movement relative to the transducer contain frequencies
different from the transmitted frequency. The change in frequency,
the Doppler shift (mf), is given by the formula, where fe is the emitted
frequency, v is the velocity of the scatters, is the angle between the
ultrasound beam and the velocity vectors in the vessel and c is the
speed of sound in the tissue.
measurement. Lateral resolution is dependent on wave
length and probe size, and decreases (lateral point-size
increases) with increasing depth (fig. 3). Consequently, the
ultrasound image is sharpest in the focus area of the transducer, and gets more blurred in larger depths [21]. Nowadays multiple focus zones overcome this problem.
Signal Attenuation
The signal intensity depends on the attenuation of the
ultrasound energy as it passes through the tissue. The
attenuation is dependent on reflection, scattering and
absorption.
Reflection. Reflection occurs at any site with different
acoustical impedance. The greater the difference in impedance, the stronger is the reflection. This occurs in the
boundary layers between tissue and bone and air, respectively, for instance against bowel gas. Ultrasound gel is
used to eliminate the layer of air between the transducer
and the skin surface, thereby avoiding reflection of the
ultrasound. However, reflection per se is not always undesirable. In fact, a reflection from boundaries within the
204
Heart Drug 2004;4:201–217
The Doppler Principle
When an object moves relative to a sound source transmitting a given frequency (e.g. an ultrasound transducer),
the frequency of the backscattered ultrasound is different
from the one emitted: the frequency is higher if the movement is towards the transducer, and lower if the movement is away from the transducer (fig. 4). This change in
frequency is called Doppler shift (¢f) and is described by
the equation:
¢f = 2 fe W v W cos /c
where fe is the emitted frequency, v is the velocity of the
moving objects (usually red blood cells in circulating
blood), is the angle between the ultrasound beam and the
vascular axis, and c is the speed of sound in the tissue (approximately 1,540 m/s). When using a pencil-probe Doppler transducer, is usually not known and may represent an
error in recording blood velocity. In duplex scanners, however, is compensated by visual angle correction, and the
velocity calculation is thus more accurate.
Small, ‘pocket Doppler’ units are usually continuous
wave (CW) Doppler velocity detectors. These include two
piezoelectric elements with overlapping ultrasound fields,
one that continuously emits ultrasound, whereas the other
is continuously detecting sound waves (fig. 5). The overlap
fields represent the area where Doppler signals may be
detected, and is termed sample volume. CW Dopplers have
no range resolution, and cannot distinguish between signals
from different vessels lying along the path of the beam.
Stranden
Pulsed wave (PW) Doppler systems are used to detect
blood velocities at specific depth. Using only one piezoelectric element, the transducer functions alternating as
transducer and receiver. The signal received is sampled at
an adjustable delay after pulse transmission. This way,
signals at a given tissue depth are selected (sample volume, fig. 5). The sample volume size is variable, and is
determined by both the characteristics of the ultrasound
beam and by the duration of the ultrasound pulse. Shorter
pulse length (shorter ‘packets’) during the transmission
phase results in a smaller sample volume, with a corresponding increase in the axial resolution [22].
The pulsing beam may, however, introduce a problem
when measuring large blood velocities, termed frequency
aliasing. To avoid aliasing, the Doppler shift must be less
than half the pulse repetition frequency, PRF (often
referred to as the Nyquist limit of the frequency). A new
transmission pulse is not allowed until the former pulse is
sampled. At larger sampling depths, the time between
transmission and gating has to be longer, because of the
longer time needed for the sound wave to reach the target
area and to return to the transducer. In turn, this increases
the time between the pulses, i.e. reduces PRF, and thereby
reduces the limit where aliasing occurs. To correctly
detect very high velocities at large depths, for instance in
mitral stenoses, it may be necessary to switch to CW to
avoid aliasing problems.
Doppler Flowmeter
In Doppler flowmeters, the mean blood velocity (cm/s;
the ‘mean blood velocity’ refers to the temporal average of
the mean velocity vector trace) is multiplied by the internal cross-sectional area (cm2) to obtain the calculated
blood flow (ml/min). This calculation is made automatically by some flowmeters by a multirange-gated ultrasonic
beam where the diameter is derived, or the flow is
obtained semi-automatically with an on-line analogue calculation unit where internal diameter is manually inserted. In both techniques, the accuracy of the calculated
volume flow is highly dependent upon the correct determination of the diameter because the error is squared at
the calculation (Q = v W  W r2), where v is velocity and r the
internal radius of the vessel. For example, if the radius of
an artery of 3.0 mm is erroneously taken as 3.5 mm, the
calculated blood flow is overestimated by 35%.
To minimize errors associated with angle insonation
and vessel diameter, the transducer may be fixed to the
vessel wall with a special cuff [23], or by customizing special transducers with inbuilt angle correction (fig. 6).
Doppler flowmetry has the advantages over the electro-
Methods to Evaluate Vascular
Reconstruction
Fig. 5. Schematic illustration of CW, PW and multirange-gated PW
Doppler techniques. CW Dopplers contain two piezoelectric crystals:
one that continuously emits ultrasound while the other continuously
detects ultrasound echoes from the tissue. The overlap field is the
location where Doppler signals may be detected, the ‘sample volume’
(SV). The sample volume is smaller in PW Dopplers, and the SV
depth may be selected. In multirange-gated PW Dopplers the SV is
sectioned, enabling identification of various Doppler spectra through
the cross-section of the vessel.
Fig. 6. Doppler ultrasound transducer made for intraoperative use
(developed by the author). The concave tip ensures good contact with
the vessel wall even without a coupling medium. The transducer is
placed at 90 ° to the vessel with the direction identifier at the top
pointing upstream. The inbuilt crystal angle correction ensures fixed
vessel insonation at 60 °.
Heart Drug 2004;4:201–217
205
Fig. 7. The principle of transit time flowmetry (A). Ultrasound is
alternatingly sent upstream (continuous line) and downstream (interrupted line). When blood is not flowing, the passage time (T) is t. The
passage time is minutely (¢t) increased when sent upstream and
reduced by the same amount when sent downstream. The magnitude
of ¢t is dependent on blood flow. In practice, the transducer crystals
are placed on the same side and the ultrasound beam is reflected at a
mirror (dual pass; B).
magnetic flowmeter of requiring no zeroing, and it is virtually unaffected by the vessel wall provided no air is
entrapped within the wall material, for instance at newly
inserted expanded PTFE and Dacron grafts. This problem might be reduced by gently squeezing the PTFE graft
between two fingers for 2–5 min, enabling blood to penetrate the porous graft material [24].
Transit Time Flowmeter
The most recently developed technique is the transit
time flowmeter, also using ultrasound, but not the Doppler principle. This method is based on the fact that ultrasound traveling against the bloodstream will take a longer
time than when moving downstream (fig. 7A). The theoretical basis for this technique has been known for some
time, but only in the latest decade practical solutions have
been available. An example is the Medi-Stim VeriQ flowmeter (fig. 8; www.medistim.com), including a computer-
206
Heart Drug 2004;4:201–217
Fig. 8. A computer-based transit time flowmeter, Medi-Stim VeriQ.
based system with trend functions and the ability to measure peripheral vascular resistance. Transit time flowmeters are very accurate and reproducible, even at minute
flow rates, and possess high zero line stability [25, 26].
A practical transducer design is shown in figures 7B
and 9, where the two ultrasound crystals are placed on one
side of the vessel and a metal reflector on the opposite
side. This induces twofold passage of ultrasound through
the blood vessels. The downstream crystal transmits a
pulse of ultrasound to the upstream crystal. The difference in transit time propagation depends on the blood
flow. The blood flow prolongs the time for the ultrasound
beam to pass through the vessel against the bloodstream,
and a phase detector senses the time difference. A new
sound beam is sent downstream and reflected. The blood
flow decreases the transit time because the transmitting
medium is also moving downstream. The process is then
reversed several hundred times a second. Since the position of the crystals is fixed, the difference in transit time
between the two directions can be related directly to blood
Stranden
Fig. 9. The two piezoelectric crystals of
transit time transducers are placed within
the housing on one side of the vessel, with a
metal reflector on the opposite side. Transducer for small-caliber vessels (A), e.g. coronary bypasses, and for large vessels (B), e.g.
aorta, are shown. C Measurement of blood
flow in a saphenous vein coronary bypass.
flow. An extremely sensitive and stable detecting device is
required to measure the very small difference between
the transit times, which is in the order of picoseconds
(10 –12 s), but the method has the advantage that the flow
direction, as well as the volume flow, is indicated. A prerequisite for a correct estimate of blood flow is even
insonation of the cross-section of the vessel. On the other
hand, the difference in the transit times is only dependent
on the moving particles in the vessel, thus making the
measurements independent of the inner diameter.
At our hospital, the use of intraoperative flowmetry
has been implemented in peripheral vascular procedures
for about 35 years. In this period, all the described techniques have been employed: In the 60s and 70s, electromagnetic flowmetry was used, in the 80s ultrasound
Doppler technique with custom-made peroperative transducers, and in the 90s we shifted to transit time flowmetry, which we regard as the method of choice for peroperative blood flow measurement.
Examples of Flowmetry in Peripheral Vascular
Disease
Detection of Side Branches during ‘in situ’ Bypass Surgery. During in situ bypass surgery, the side branches of
the great saphenous vein must be interrupted to prevent
them from becoming arteriovenous fistulae after the reconstruction. Following completion of the proximal and
distal anastomoses, vessel clamps are removed and blood
flow through the bypass is established. Major tributaries
of the vein can be visually identified and ligated without
difficulty. However, smaller branches often have to be
detected by other means, mainly intraoperative flowmetry, in order to avoid major surgical dissection. It is
Methods to Evaluate Vascular
Reconstruction
important to locate these residual fistulae since they may
cause graft failure [27].
The detection of arteriovenous connections has a qualitative aim; exact blood flow values are less relevant. A
flow probe of appropriate size is placed around the vein,
just below the proximal anastomosis. When using transit
time flowmetry, either sterile saline or ultrasound gel is
used to maintain acoustic coupling between the probe and
the measuring site. Pulsatile flow curves will immediately
appear on the display (fig. 10). Manually or by means of
an atraumatic clamp, the vein is occluded successively at
different sites along its course, from the transducer to the
distal anastomosis.
A reverberating flow profile, indicating no net blood
flow (fig. 10, position 1), is found when there is no leakage
flow between transducer and the compression site. If flow
is detected during clamping, we can expect to find an
open side branch proximal to the site of clamping (fig. 10,
position 2), which can subsequently be ligated.
Intraoperative Functional Evaluation of a Vascular Reconstruction. As distal femorotibial bypass grafting has
become more common, supplementary methods are necessary for intraoperative control. The primary aim is to
obtain information on the prognosis for the immediate
result of the reconstruction. Several studies have shown
that peroperative flow values have prognostic values [1, 3,
5]. The risk of early postoperative occlusion is significantly increased if the basal blood flow after femoropopliteal
reconstruction is less than 100 ml/min or the papaverineinduced flow (intra-arterial injection of 40 mg papaverine) is less than 200 ml/min (fig. 11). The effect of papaverine is reduced if the surgery is performed under epidural
anesthesia, since basal flow is already increased.
Heart Drug 2004;4:201–217
207
A
B
Fig. 10. Procedure for the detection of side branches during in situ
bypass surgery. During the examination, the finger is compressing
the artery. If net flow is detected (Pos. 2), the finger is downstream a
side branch acting as arteriovenous fistula. This is subsequently
ligated.
Obviously, blood flow values do not necessarily provide information about anatomical aberrations due to
technical failure. Ideally, intraoperative arteriography or
B-mode scanning is performed in order to supply the surgeon with an anatomical evaluation as well.
Intra-Arterial Pressure Recording
Arteriography alone may be inadequate in the prediction of the hemodynamic significance of lesions, particularly in the aortoiliac region, as reported by Sumner and
Strandness [28]. The study noted that up to 30% of
patients who had combined aortoiliac and femoropopliteal disease did not show hemodynamic improvement
after a proximal reconstruction. This confirmed the inadequacy of arteriography alone in predicting the outcome
of a proximal arterial reconstruction.
208
Heart Drug 2004;4:201–217
Fig. 11. Mean and pulsatile blood flow curves in a successful femoropopliteal bypass. A Basal flow. B Following intra-arterial injection of
40 mg papaverine.
Most investigators have generally agreed upon that any
attempt to predict hemodynamic significance of a stenosis must be verified by direct intra-arterial pressure measurements [29]. Normally, the systolic pressure gradient
between the aorta and the common femoral artery is
below 10 mm Hg. In cases with unclear angiographic findings and a gradient below 10 mm Hg, intra-arterial papaverine is given, leading to a transient vasodilatation and
increase in flow. Administration of papaverine is designed to mimic the flow increase seen with exercise. A
systolic pressure gradient of more than 20 mm Hg is considered hemodynamically significant [30].
There is little doubt that combined intra-arterial pressure and arteriography is extremely useful for documenting the hemodynamic status of an arterial segment.
It has been suggested that these measurements must be
done when stenotic lesions are found and angioplasty is
Stranden
being contemplated [29]. Furthermore, the pressure gradient should in addition be measured after completion of
the dilatation to document the immediate result.
When reference pressure is obtained at the arm, great
care should be taken when interpreting the data. Carter
[31] and Carter and Tate [32] have shown that large systolic pressure wave amplification in the peripheral vessels
may exist, especially in body cooling. Therefore, if brachial, or even worse, radial pressure is taken to represent
central prestenotic pressure, an overestimation of that
pressure may be significant. Hence the pressure gradient
over the stenosis may be overestimated. This effect is
reduced when the body is kept warm to induce peripheral
vasodilatation.
When pressure is recorded during angiography, a ‘pullthrough’ technique should be applied, whenever possible.
The pressure transducer catheter line is introduced and
moved proximal to the obstruction to measure the preobstruction pressure. Then the catheter is pulled through
the obstruction to obtain the post-obstruction pressure.
The difference between these pressures indicates the degree of obstruction.
Measurement of Vascular Outflow Resistance
Following femorodistal bypass surgery, both limb salvage rate and graft patency are dependent upon outflow
resistance. Preoperative arteriographic visualization of
these vessels does not inform us on whether their functional capacity is adequate to keep a distal bypass open.
Studies have shown that intraoperative measurements of
distal resistance are predictive in terms of outcome of
femorodistal bypasses [33–35]. Some authors have defined a sharp cutoff point above which femorodistal grafts
are prone to reocclude [34]. Attempts have been made to
define a level of resistance above which primary amputation should be performed instead of arterial reconstruction [36], or levels where reopening of the graft can be
neglected in case reocclusion occurs in the early postoperative phase. However, factors like the nature of the fluid
infused during the measurement or the tone of the peripheral vascular tree, for example, may significantly influence the results [36, 37].
Outflow resistance (R) has been examined with different techniques, one of which is the basic pressure (P)
divided by flow (Q) measurement, which represents the
definition of vascular resistance: R = P/Q, expressed in
peripheral resistance units (mm Hg W min W ml –1). This
technique was simplified, as described by Ascer et al. [33].
A mathematical rewriting simplifies the equation to include the pressure integral when a fixed volume (V) is
Methods to Evaluate Vascular
Reconstruction
Fig. 12. Outflow resistance measurement by the pressure integral
technique. The mathematical basis is shown in the equation. Vascular resistance (R) is by definition the mean pressure (P) divided by
mean blood flow (Q), for instance the pressure increase when a volume is injected by a syringe. This expression may be integrated as
shown in the middle equation. However, as the time integral of flow
is equal to the volume injected in that time period (V), the middle
equation is reduced to constitute the pressure integral divided by a
fixed volume. In the unit shown below (‘PresRes Monitor’, developed by the author) the vascular outflow resistance is obtained by
pressure recording during infusion, subtracted of backpressure (zeroing), sampled during a fixed infusion time (infusion flow 100 ml/
min) and automatically integrated during that time. When the proximal part of the artery is occluded (‘Occl.’), the distal runoff resistance
is recorded. When the occlusion is removed, some of the fluid may be
squeezed also in proximal direction, and the total vascular resistance
distal to a vascular ‘end-to-side’ graft is recorded.
injected into the distal artery (fig. 12). In this technique,
flowmetry is not needed as a special pressure integration
unit is used.
Noninvasive techniques examining outflow vessels
preoperatively have also been presented. The most promising is the pulse-generated runoff, where a pressure cuff is
placed around the calf or thigh, and Doppler examination
of the calf (or foot) vessels is performed after rapid repeti-
Heart Drug 2004;4:201–217
209
there are a few disadvantages with the method. Usually
only the graft and the distal anastomosis are visualized,
which is unfortunate since technical defects may also be
located at the proximal anastomosis. The arteriograms are
normally taken in one plane only, and small defects may
be hidden in the contrast material. The method includes
injection of contrast material, which may lead to allergic
reactions, and may be nephrotoxic.
Fig. 13. Schematic presentation of the noninvasive pulse-generated
runoff technique for evaluating outflow vessels prior to surgery. A
proximal occlusion cuff prevents interference from the proximal
arterial pulse. A pressure cuff placed at calf level is intermittently
inflated to produce artificial pressure/flow waves, which may be
detected in the distal branches when the artery is open (lower left
curve). The pedal arch is examined by manually compressing
(+Comp.) the leg arteries as described by Scott et al. [40].
tive inflations of the cuff [38, 39]. The method has a semiquantitative scoring system. With the technique, a standardized artificial arterial pressure wave is created in the
calf and detected in the leg/foot arteries by Doppler ultrasound technique. The latest refinement of the procedure
includes a pedal arch patency test (fig. 13). This test,
applied to all three leg arteries, may predict the likelihood
for a femorodistal bypass to remain open, and could be of
help to select patients where a primary amputation should
be considered [40].
Anatomical Tests
Intraoperative Arteriography
Intraoperative arteriography is regarded as the gold
standard in the control of infrainguinal reconstruction,
either by conventional arteriography or digital subtraction arteriography. It has been shown that technical
defects may be observed in more than 20% of the cases
[41]. Due to the direct visualization of the vessels, interpretation of the findings is generally easy. By detecting
defects already during the operation, they can be corrected, thereby improving the patency rate. However,
210
Heart Drug 2004;4:201–217
Magnetic Resonance Angiography (MRA)
For pre- and postoperative evaluations, MRA has become a noninvasive alternative to arteriography. This
technique is advantageous as it does not require exposure
to ionizing radiation and arterial catheterization. Different MRA techniques have been developed, with the latest,
contrast-enhanced MRA being the most promising for the
evaluation of the aortoiliac and infrainguinal vasculature.
The tendency of overestimating the degree of stenosis
may still be a problem, though.
Angioscopy
Following the introduction of modern fiberoptic angioscopes, the graft and the anastomotic area can be
visualized. Angioscopy may be used to detect residual valvular flaps, which may be retained following in situ vein
bypass grafting (fig. 14). The method can also be used
with endovascular closure of venous side branches, and in
evaluating venous valves. In some areas, angioscopy is
superior to arteriography. Neville et al. [42] found in an in
vivo model that planar arteriography in one projection
identified 60% of the arterial intimal flaps, while angioscopy and intravascular ultrasound visualized all of them.
Comparing angioscopy with arteriography, the latter
method is especially inferior in identifying thrombi;
sometimes arteriography underestimates or misses mural
or retained thrombi and debris [43].
A disadvantage of angioscopy is the lack of visualization of the outflow arteries including the pedal arch,
whereas angiography visually includes these arteries. Furthermore, only anatomical details may be shown, and no
impression of the peripheral vascular resistance is obtained. Sometimes clear visualization is difficult due to
bleeding from side branches, and the method may be
technically difficult to perform. The equipment is expensive, especially disposable angioscopes, and there may be
difficulties with cleaning and sterilization.
Color-Flow Duplex (CD) Scanning
CD scanning is a direct noninvasive technique and
includes three modalities: real-time B-mode (brightness
Stranden
mode) image, Doppler velocity spectrum (usually PW) and
color-coded flow information of blood velocity superimposed on the B-mode image. The technique provides both
anatomical and physiological information, and has become the key instrument in vascular laboratories.
Pre- and Postoperative Evaluation. Color-coding has
the advantage over plain duplex (B-mode + spectral
Doppler) that it enables rapid identification of arteries,
and specifically areas of stenoses and obstructions with
flow disturbances. Atherosclerotic plaques are identified
by the presence of calcification or increased echogenicity.
Stenoses can be detected using different criteria, such as
peak systolic velocity (PSV) within the narrowing, the
ratio of PSV in the stenosis and PSV in a normal segment
close to it (PSVR), end-diastolic velocity or, more subjectively, by noting the degree of spectral broadening. Several studies indicate a high degree of accuracy in the assessment of the aortoiliac and femoropopliteal arteries, but
poorer accuracy in distal arteries [44]. Most centers use a
PSV ratio 12.0 as a criterion of stenosis to detect a stenosis with a diameter reduction 150%. These studies have
challenged arteriography as the golden standard in the
evaluation of lower-extremity atherosclerosis, especially
in the aortoiliac and femoropopliteal arteries. Consequently, an increasing number of hospitals now base the
selection of therapy on CD investigations. For planning
femorodistal reconstructions, most surgeons will find duplex scanning not optimal and regard arteriography a prerequisite.
Deep vessels, low blood velocities, calcified plaques
with acoustic shadows and bowel gas frequently reduce
the quality of the CD scanning, which is thus sometimes
inconclusive. Software enhancement, for example power
mode Doppler based on signal intensity rather than the
Doppler shift, increases sensitivity, and may be used to
improve locating vessels with low blood flow states or
improve delineation of hypoechoic plaques. Furthermore,
echo enhancers (ultrasound contrast), administered intravenously, may significantly improve the investigation in
problem areas, e.g. in arteries of the pelvis, distal thigh
(Hunter’s canal) and calf.
Duplex scanning provides tissue images in multiple
planes on the operating table, together with a velocity profile analysis for the evaluation of the hemodynamic
results. Color codification of the blood velocity placed on
the tissue image provides additional global, two-dimensional information on the flow patterns in the vessels,
anastomoses and grafts [45]. Furthermore, anatomical
defects, e.g. loose intimal flaps, strictures, intraluminal
thrombi, loose debris, residual stenosis and misplaced
Methods to Evaluate Vascular
Reconstruction
Fig. 14. Angioscopy images from venous (A–C) and arterial (D)
investigations. A Incompetent venous valve with floppy cusps that
do not approximate during Valsalva maneuver. The same valve after
successful angioscopically guided external valvuloplasty during
opening (B) and closing (C). The images are retained from a video
sequence, hence the somewhat reduced image quality. For the benefit
of clarity, the edges of the valve cusps are artificially outlined. D A
residual intimal flap after arterial reconstruction.
sutures, may be identified [46] (fig. 15). Flow abnormalities found with Doppler spectral analysis or Doppler color-coding helps to identify significant lesions [47]. According to published reports on the intraoperative use of
duplex ultrasound to assess reconstructive vascular surgery, defects are found in 20–30%. Only 5–10% of these
are judged as significant and consequently corrected. The
indications for vessel reentry are subjective and empirical. The decision to revise a reconstruction is principally
based on the surgeon’s judgment that the residual anatomic defect or hemodynamic disturbance may cause
complications.
Intravascular Ultrasound
For endovascular procedures intravascular (intraluminal) ultrasound may be the method of choice, since it can
visualize the composition of the lesion, and thereby help
to select the proper treatment modality. Furthermore,
Heart Drug 2004;4:201–217
211
Fig. 15. An intimal defect (arrow) protruding into the lumen following carotid endarterectomy detected by color duplex scanning
(A). At reintervention, an intimal flap of 2 !
7 mm was identified and removed (B). C Bmode image of a residual intimal flap detected following carotid endarterectomy (arrow). The wide arrow indicates the direction
of blood flow. After revising the defect, a
new examination identified a fresh thrombus (D, arrow) at the site where an arterial
clamp was applied during the first revision.
Consequently, a second reopening was necessary. The images are kindly provided by
O.D. Saether (A, B) and T. Dahl (C, D),
Trondheim University Hospital, Trondheim, Norway.
Fig. 16. Suggested single introduction, therapeutical IVUS unit with
a PTA balloon that might be available in the future. This unit combines the description of intraluminal anatomy (IVUS), physiology
[pressure gradient (P2-P1) and Doppler ultrasound] and therapy
(PTA with/without stent) in one introduction procedure.
residual plaques or flaps can be detected, and a sufficient
stent wall contact can be confirmed [48]. The technique
will earn more attention when combined with therapy
constitution. A single introduction, therapeutic IVUS unit
with a PTA balloon, which could be reality in the future, is
212
Heart Drug 2004;4:201–217
hypothesized in figure 16. As microcatheter technology is
being refined, one might expect that addition of dual pressure sensors and a Doppler ultrasound transducer at the
tip should not raise the catheter cost significantly. With
the pressure sensors, indicated as P1 and P2, the pressure
gradient is measured continuously to evaluate the immediate effect of angioplasty. Before dilatation, the estimated degree of stenosis is however overestimated, as the
catheter occupies a fraction of the cross-section area of the
narrowed lumen. These measurement entities are already
available on separate units: miniaturized pressure sensors
are available in angiography guide wires (e.g. Cardiometrics WaveWire, Endosonics, San Diego, Calif., USA), as is
Doppler ultrasound transducers (e.g. Cardiometrics FloWire, Endosonics). Combined IVUS/PTA catheters are
already available (e.g. MegaSonics F/X and OTW PTCA
catheters, Endosonics).
Three-Dimensional Mode
Vascular disorders are essentially three-dimensional
and should logically be visualized that way, especially in
regard to vascular malformations and the quantification
of atherosclerotic plaques [49, 50]. Current technology
has been used for three-dimensional imaging of larger
Stranden
organs like the heart and kidney, and volume estimation
of abdominal organs [51, 52]. In particular, three-dimensional echocardiography has proved to render important
additional clinical information [53–57].
Three-/Four-Dimensional Representation
A two-dimensional ultrasound image is composed of
X W Y picture elements, or pixel. This concept can be
extended to the imaging of three-dimensional objects. The
depth, or Z dimension, can be modeled by stacking a series
of two-dimensional planes in depth. The two-dimensional
picture elements are then interpolated to become cubes.
These cubes, which form the three-dimensional image, are
called volume elements or voxels. The volume representation is called a cuberille. When volumetric information is
correlated with time to form the fourth dimension, every
voxel element has variations over time. Representation in
four dimensions is particularly used when time dynamics
plays an essential role, e.g. displaying the heart or cardiovascular structures over a cardiac cycle.
At present, the process of making three-dimensional
images based on ultrasonography is divided into four
steps (fig. 17): data acquisition and filtering; volume generation; data processing, and visualization.
Several methods are available to acquire two-dimensional images to be transformed into three-dimensional
data. The region of interest may be insonated by movement of the probe, by rotating, tilting or linear translation.
When a position sensor system is applied, for instance
based on magnetic tracking, the transducer may be moved
about freely to generate the set of two-dimensional images
that are transformed into the three-dimensional domain
[58]. In the future, two-dimensional array transducers are
probably employed to a larger degree. Very high processing speeds and capacities are however needed for realtime three-dimensional display. The acquisition may involve respiration and ECG triggering to reduce movement artifacts.
During three-dimensional data acquisition and image
processing, the data undergo several transformation and
filtering operations. The first post-processing step is to
transform the two-dimensional video-grabbed data into a
cubic representation of the image volume by interpolating
neighboring pixels in the two-dimensional video-capture.
The next post-processing step, orientation, refers to the
interactive process of reviewing the data when ‘viewing’
windows are opened for examining the volume. This
viewing function is the essential tool for the positioning of
viewing planes in relation to objects or structures that are
of clinical interest.
Methods to Evaluate Vascular
Reconstruction
A
B
C
D
Fig. 17. Ultrasonic 3D acquisition (A) and post-processing (B–D).
A A 3D image volume is created by translation of the ultrasound
transducer, by rotation, linear translation or fan sweep. When a position sensor system is applied, the transducer may be moved about
freely to generate a set of 2D images. In the future, 2D matrix transducers are used most probably. B Volume generation: a series of 2D
sections are placed one behind the other to form an image volume. A
reference image is chosen within this series where important clinical
information is displayed. C, D Visualization of the region of interest
within the image volume is done with virtually cut planes (viewing
planes) in any direction and location within the volume. These cut
planes may be freely moved and rotated at any angle. The cut planes
are slices in the image volume and visualized as images either based
on tissue or flow data (D).
Heart Drug 2004;4:201–217
213
Fig. 18. Views of the proximal anastomosis of a patient with a
reversed vein femoropopliteal bypass to the common femoral artery
(cfa) applying the 3D gradient-shading technique. The 3D image is
generated from the longitudinal reference image at the top and subsequently underlying parallel planes (in depth). The arrows point to the
suture line of the anastomosis and what is supposed to be a wall stricture.
The final phase of post-processing is the computation of
three-/four-dimensional images containing information
on tissue surfaces, their orientation and distance from the
viewing plane. Surface points that are more distant from
the observer will become darker than points that are closer
to the observer. Our studies on vascular three-dimensional
processing [59] indicate that gradient shading is an applicable and robust technique. In gradient shading, the orientation of surface together with distance determines the
gray value, providing images with visual appearance correlating well with normal photographical techniques.
Figure 18 depicts the proximal anastomosis of a patient
with a reversed vein femoropopliteal bypass applying the
three-dimensional gradient-shading technique.
214
Heart Drug 2004;4:201–217
Color Doppler velocity or power energy mode data
have been used to isolate vascular structures from gray
scale tissue images, thereby obtaining clearer vessel anatomy [59–61]. The use of velocity or power mode data has a
twofold advantage over gray-scale-generated images.
Firstly, a major difficulty in volume data visualization is
that the majority of volume elements do not contain clinically relevant data. Extraction of clinically useful information may therefore be difficult other than displaying a
two-dimensional slice. Using flow information reduces
this problem. Secondly, eliminating gray scale and instead
using flow data for surface rendering also improves visualization of vascular structures (fig. 19). This is especially important when understanding complex spatial relationships.
Three-dimensional representation of the vasculature
may become a valuable adjunct to conventional twodimensional imaging. An advantage of the three-dimensional representation lies in the fact that orientations are
preserved, enabling descriptive data, not definable by
two-dimensional formats or a verbal report, to be conveyed rapidly and effectively to the clinician. This may
enable less-experienced physicians to interpret complex
images and to make diagnostic decisions that currently
require specialists in sonography. The effectiveness of this
transfer of information, however, depends on how realistic the three-dimensional images appear.
Mathematical modeling may be applied to the acquired dataset to calculate the volume of the structures
[51]. The Vingmed EchoPac 3D (www.geultrasound.com)
facilitates different acquisition modes and is computer
platform independent. Interactive contouring does the
volume modeling, and the volume calculations have proven very accurate. An example is shown in figure 20.
Our experience shows that three-/four-dimensional ultrasound is superior to two-dimensional imaging in communicating volumetric information, and the use of the
ultrasound power energy mode may overcome problems
related to poor B-mode visualization. Clinically, threedimensional imaging may become important in relation
to arterial reconstructions performed without angiography, e.g. carotid endarterectomies or even beating heart
coronary surgery in the future.
Cost-Effectiveness
In the future, it is likely that health economy will be
even more constrained than today. Cost-effectiveness of
intraoperative methods will therefore have to be evaluated thoroughly before they are introduced into the clinical
routine. The selection of methods will be based on the
Stranden
19
Fig. 19. Sections from rotation of a 3D
image of a carotid bifurcation. The images
are based on power mode Doppler data. Tissue data are suppressed by reducing the tissue gain level. The arrows point to irregularities of the flow contour, coinciding with stenotic parts of the vessels. cc = Common
carotid artery; ic = internal carotid artery;
ec = external carotid artery.
Fig. 20. Calculation of the volume of a
femoropopliteal bypass aneurysm using
Vingmed Echopac 3D. The volume is the
demarcation of the open flow section of the
aneurysm. The volume outside this contour,
underneath the vessel wall (interrupted line),
is thrombotic material. The lighter section
above the flow-generated volume is a split in
the thrombus, verified at surgical removal of
the aneurysm (arrow).
Methods to Evaluate Vascular
Reconstruction
20
Heart Drug 2004;4:201–217
215
number of reconstructions performed by the institution
each year, and the economical situation of the hospital.
Furthermore, the procedure must be easy to use, and
should not prolong the operating time unduly.
The costs of the procedures must be weighed against
the possibility of leaving a less satisfactory reconstruction.
In the long run, this could lead to graft occlusion with
amputation as a result in patients with critical limb ischemia [62, 63]. Initially the cost of amputation is similar to
vascular reconstructions, but the patient with an amputation may be a burden for the society for many years. Consequently, efforts to improve the patency rate of reconstructions for critical limb ischemia can be regarded as
good economy.
Future
Angiography is one of the cornerstones in the intraoperative diagnostics in critical limb ischemia, and in
many centers regarded as part of the minimum facilities
in the management of these patients. Angioscopy still has
to be evaluated as far as the efficiency of detecting significant failures are concerned. These methods will probably
supplement each other in the near future.
Further refinements in three-dimensional imaging of
the vessels are anticipated, especially in the ultrasound
techniques. Therapeutic IVUS catheters with pressure
transducers and Doppler ultrasound for PTA and stent
insertion probably also emerge. Real-time three-dimensional ultrasound scanners require a fast computer kernel
and improved transducer technology, probably a crystal
matrix design. With such technology, a three-dimensional
representation of blood flow and flow vectors may enhance our perception of local flow conditions and morphological changes in diseased states. Along that chain of
evolutions, we hope that color maps for energy dissipation
are developed and implemented to directly locate and
visualize the hemodynamic importance of arterial obstructions.
Resistance measurements will probably have a more
important place in this area in the future. Patients who
end up with an amputation following vascular surgery
have often undergone several attempts of revascularization prior to amputation. Thus, a key problem is to decide
when to do a primary amputation, or when to give the
patient the chance of reconstruction, and when to refrain
from a second attempt in case reocclusion occurs. Finally,
resistance measurements can perhaps give us some idea
when it will be feasible to perform a femorodistal bypass
graft to the leg arteries. Reliable data concerning cutoff
values are warranted.
Two basic functional tests are going to persist: flowmetry and intra-arterial pressure measurements. Of the flowmeters, transit time technology will be chosen because of
its accuracy and ease of handling. When combined, these
measurements may constitute outflow resistance.
Acknowledgment
Parts of this paper are based on a book chapter [64].
References
1 Dedichen H: Hemodynamics in arterial reconstructions of the lower limb. I. Blood flow. Acta
Chir Scand 1976;142:213–220.
2 Stranden E, Myhre HO: Intra-operative diagnostics in critical limb ischaemia. Critical Ischaemia 1994;4:44–52.
3 Dedichen H: The papaverine test for blood
flow potential of ilio-femoral arteries. Acta
Chir Scand 1976;142:107–113.
4 Hickey NC, Shearman CP, Crowson MC, et al:
Iloprost improves femoro-distal graft flow after
a single bolus injection. Eur J Vasc Surg 1991;
5:19–22.
5 Ihlberg LHM, Albäck NA, Lassila R, Lepäntalo
M: Intraoperative flow predicts the development of stenosis in infrainguinal vein grafts. J
Vasc Surg 2001;34;269–276.
6 Cappelen C Jr, Efskind L, Hall KV: Electromagnetic flowmeter measurements of the
blood flow in the ascending aorta during cardiac surgery. Acta Chir Scand Suppl 1966;
356B:129–133.
216
7 Cappelen C Jr, Hall KV: Electromagnetic
blood flowmetry in clinical surgery (review).
Acta Chir Scand Suppl 1967;368:3–27.
8 Cronestrand R, Ekestrom S, Hambraeus G:
The value of blood flow measurements in acute
arterial surgery. Scand J Thorac Cardiovasc
Surg 1969;3:48–51.
9 Cappelen C Jr, Hall KV: Intra-operative blood
flow measurements with electromagnetic flowmeter. Prog Surg 1970;8:102–123.
10 Semb GS, Cappelen C Jr, Hall KV, Efskind L:
Postoperative aortic regurgitation related to
peroperative blood flowmetry in ball valve replacement. Scand J Thorac Cardiovasc Surg
1970;4:25–30.
11 Hall KV, Fjeld NB: Peroperative assessment of
run-off by electromagnetic flowmetry. Scand J
Clin Lab Invest Suppl 1973;128:185–188.
12 Foxworthy JV, Monro JL, Lewis B: The response to papaverine in coronary artery bypass
graft flows. J Cardiovasc Surg 1985;26:439–
442.
Heart Drug 2004;4:201–217
13 Walpoth BH, Bosshard A, Genyk I, Kipfer B,
Berdat PA, Hess OM, Althaus U, Carrel TP:
Transit-time flow measurement for detection
of early graft failure during myocardial revascularization. Ann Thorac Surg 1998;66:1097–
1100.
14 Walpoth BH, Bosshard A, Kipfer B, Berdat PA,
Althaus U, Carrel TP: Failed coronary artery
bypass anastomosis detected by intraoperative
coronary flow measurement. Eur J Cardiothorac Surg 1998;14(suppl 1):S76–S81.
15 D’Ancona G, Karamanoukian HL, Salerno
TA, Schmid RN, Bergsland J: Flow measurement in coronary surgery. Heart Surg Forum
1999;2:121–124.
16 Geddes LA, Baker LE: Principles of Applied
Biomedical Instrumentation, ed 2. New York,
Wiley, 1975.
17 Sykes MK, Vickers MD, Hull CJ: Principles of
Clinical Measurement. Oxford, Blackwell
Scientific, 1981.
Stranden
18 Roberts VC: Haematocrit variations and electromagnetic flowmeter sensitivity. Biomed Eng
1969;4:408–412.
19 Dennis J, Wyatt DG: Effect of hematocrit value upon electromagnetic flowmeter sensitivity.
Circ Res 1969;24:875–886.
20 Lundell A, Bergqvist D: Intraoperative flow
measurements in vascular reconstruction. Ann
Chir Gynaecol 1992;81:187–191.
21 Angelsen BAJ: Ultrasound imaging – Waves,
signals and signal processing. Norwegian University of Science and Technology 2000, pp.
1416, www.ultrasoundbook.com.
22 Hennerici MG, Neuerburg-Heusler D: Vascular Diagnosis with Ultrasound. Stuttgart,
Thieme, 1997.
23 Beard JD, Evans JM, Skidmore R, Horrocks
M: A Doppler flowmeter for use in theatre.
Ultrasound Med Biol 1986;12:883–889.
24 Rostad H, Grip A, Hall C: Blood flow measurement in PTFE grafts. J Cardiovasc Surg (Torino) 1987;28:262–265.
25 Albäck A, Mäkisalo H, Nordin A, Lepäntalo
M: Validity and reproducibility of transit time
flowmetry. Ann Chir Gynaecol 1996;85:325–
331.
26 Laustsen J, Pedersen EM, Terp K, Steinbruchel
D, Kure HH, Paulsen PK, Jorgensen H, Paaske
WP: Validation of a new transit time ultrasound flowmeter in man. Eur J Vasc Endovasc
Surg 1996;12:91–96.
27 Donaldson MC, Mannick JA, Whittemore AD:
Causes of primary graft failure after in situ
saphenous vein bypass grafting. J Vasc Surg
1992;15:113–120.
28 Sumner DS, Strandness DE Jr: Aortoiliac reconstruction in patients with combined iliac
and superficial femoral occlusion. Surgery
1977;28:348–355.
29 Strandness DE Jr: Transluminal angioplasty: A
surgeon’s viewpoint. Am J Roentgenol 1980;
135:998–1000.
30 Thiele BL, Bandyk DF, Zierler RE: A systematic approach to the assessment of aortoiliac disease. Arch Surg 1983;18:477–485.
31 Carter SA: Effect of age, cardiovascular disease, and vasomotor changes on transmission
of arterial pressure waves through the lower
extremities. Angiology 1978;29:601–616.
32 Carter SA, Tate RB: The effect of body heating
and cooling on the ankle and toe systolic pressures in arterial disease. J Vasc Surg 1992;16:
148–153.
33 Ascer E, Veith FJ, Morin L, et al: Quantitative
assessment of outflow resistance in lower extremity arterial reconstruction. J Surg Res
1984;37:8–15.
34 Ascer E, Veith FJ, White-Flores SA, et al:
Intraoperative outflow resistance as a predictor
of late patency of femoropopliteal and infrapopliteal bypasses. J Vasc Surg 1987;5:820–
827.
35 Beard ID, Scott DIA, Skidmore R, et al: Operative assessment of femorodistal bypass grafts
using a new Doppler flowmeter. Br J Surg
1989;76:925–928.
Methods to Evaluate Vascular
Reconstruction
36 Parvin SD, Evans DH, Bell PRF: Peripheral
pressure measurement in the assessment of severe peripheral vascular disease. Br J Surg
1985;72:751–753.
37 Wahlberg E, Line PD, Olofsson P, Swedenborg
J: Infusion methods for determination of peripheral resistance – Influence of infused medium and back pressure. Ann Vasc Surg 1994;
8:172–178.
38 Beard JD, Scott DIA, Evans JM, Skidmore R,
Horrocks M: Pulse-generated runoff: A new
method of determining calf vessel patency. Br J
Surg 1988;75:361–363.
39 Davies AH, Horrocks M: Outflow resistance
measurements. Ann Chir Gynaecol 1992;81:
183–186.
40 Scott DJA, Horrocks EH, Kinsella D, Horrocks
M: Preoperative assessment of the pedal arch
using pulse generated runoff and subsequent
femorodistal outcome. Eur J Vasc Surg 1994;8:
20–25.
41 Myhre HO, Kordt KF, Stranden E: Pre- and
perioperative angiography and clinical physiological measurements. Acta Chir Scand 1987;
538:132–138.
42 Neville RF, Yasuhara H, Watanabe B, et al:
Endovascular management of arterial intimal
defects: An experimental comparison by arteriography, angioscopy, and intravascular ultrasonography. J Vasc Surg 1991;13:496–502.
43 Segalowitz J, Grundfest WS, Treiman RL, et al:
Angioscopy for intraoperative management of
thromboembolectomy. Arch Surg 1990;125:
1357–1361.
44 Koelemay MJW, Hartog D, Prins MH, et al:
Diagnosis of arterial disease of the lower extremity with duplex ultrasonography. Br J Surg
1996;83:404–409.
45 Sæther O, Mathisen S: Intraoperative Doppler.
Ann Chir Gynaecol 1992;81:176–177.
46 Myhre HO, Saether OD, Mathisen SR, Angelsen BAJ: Color coded duplex scanning during
carotid endarterectomy; in Greenhalgh RM,
Hollier LH (eds): Surgery for Stroke. Philadelphia, Saunders, 1993, pp 253–258.
47 Bandyk DF, Kaebnick HW, Adams MB,
Towne JB: Turbulence occurring after carotid
bifurcation endarterectomy: A harbinger of residual and recurrent stenosis. J Vasc Surg 1988;
7:261–274.
48 Pandian NG, Kreis A, Weintraub A, Kumar R:
Intravascular ultrasound assessment of arterial
dissection, intimal flaps, and intraarterial
thrombi. Am J Card Imaging 1991;5:72–77.
49 Delcker A, Diener HC: Quantification of atherosclerotic plaques in carotid arteries by threedimensional ultrasound. Br J Radiol 1994;67:
672–678.
50 Palombo C, Kozakova M, Morizzo C, Andreuccetti F, Tondini A, Palchetti P, Mirra G,
Parenti G, Pandian NG: Ultrafast three-dimensional ultrasound: Application to carotid
artery imaging. Stroke 1998;29:1631–1637.
51 Gilja OH, Thune N, Matre K, Hausken T, Odegaard S, Berstad A: In vitro evaluation of threedimensional ultrasonography in volume estimation of abdominal organs. Ultrasound Med
Biol 1994;20:157–165.
52 Gilja OH, Hausken T, Berstad A, Odegaard S:
Measurements of organ volume by ultrasonography (review). Proc Inst Mech Eng [H] 1999;
213:247–259.
53 Pandian NG, Roelandt J, Nanda NC, Sugeng
L, Cao QL, Azevedo J, Schwartz SL, Vannan
MA, Ludomirski A, Marx G, et al: Dynamic
three-dimensional echocardiography: Methods
and clinical potential. Echocardiography 1994;
11:237–259.
54 De Castro S, Yao J, Fedele F, Pandian NG:
Three-dimensional echocardiography in ischemic heart disease. Coron Artery Dis 1998;9:
427–434.
55 De Castro S, Yao J, Pandian NG: Threedimensional echocardiography: Clinical relevance and application. Am J Cardiol 1998;81:
96G–102G.
56 Kardon RE, Cao QL, Masani N, Sugeng L,
Supran S, Warner KG, Pandian NG, Marx
GR: New insights and observations in threedimensional echocardiographic visualization
of ventricular septal defects: Experimental and
clinical studies. Circulation 1998;98:1307–
1314.
57 Acar P, Laskari C, Rhodes J, Pandian N, Warner K, Marx G: Three-dimensional echocardiographic analysis of valve anatomy as a determinant of mitral regurgitation after surgery for
atrioventricular septal defects. Am J Cardiol
1999;83:745–749.
58 Gilja OH, Hausken T, Olafsson S, Matre K,
Odegaard S: In vitro evaluation of three-dimensional ultrasonography based on magnetic
scanhead tracking. Ultrasound Med Biol 1998;
24:1161–1167.
59 Stranden E, Slagsvold C-E, Morken B, Alker
HJ, Bjørdal J: Three-dimensional ultrasound
visualization of peripheral vessels; in Greenhalgh RM (ed): Vascular Imaging for Surgeons.
London, Saunders, 1995, pp 71–80.
60 Pretorius DH, Nelson TR, Jaffe JS: 3-dimensional sonographic analysis based on color flow
Doppler and gray scale image data: A preliminary report. J Ultrasound Med 1992;11:225–
232.
61 Li X, Shiota T, Delabays A, Teien D, Zhou X,
Sinclair B, Pandian NG, Sahn DJ: Flow convergence flow rates from 3-dimensional reconstruction of color Doppler flow maps for computing transvalvular regurgitant flows without
geometric assumptions: An in vitro quantitative flow study. J Am Soc Echocardiogr 1999;
12:1035–1044.
62 Mackey WC, McCullough JL, Conlon TP, et al:
The costs of surgery for limb-threatening ischemia. Surgery 1986;99:26–35.
63 Raviola CA, Nichter LS, Baker JD, et al: Cost
of treating advanced leg ischemia. Arch Surg
1988;123:495–496.
64 Stranden E: Functional and anatomical tests to
evaluate graft patency after vascular surgery
procedures; in D’Ancona G, Bergsland J, Karamanoukian H, Salerno T (eds): Intraoperative
Graft Patency Verification in Cardiac and Vascular Surgery. Armonk, Futura Publishing,
2001.
Heart Drug 2004;4:201–217
217